The DHFR origin consists of a zone of >22 inefficient initiation sites scattered throughout the 55 kb spacer between the convergently-transcribed DHFR and 2BE2121 genes. However, some sites (ori-[unreadable] and ori-?) are clearly preferred. We have shown that in loco deletion of the most active initiation sites from the DHFR origin, or even the central 40-kb core that encompasses >90% of active sites, does not suppress initiation in the remainder of the spacer. In fact, initiation frequency actually increases. Thus, master replicators do not reside in the origin itself. However, an active DHFR promoter is required for maximum origin activity and, in its absence, replication initiates in the now-inactive gene. Furthermore, the 3' processing signals of the gene prevent inactivation of the origin by read-through transcription. We propose that mammalian genomes are peppered at intervals of <1 kb with degenerate replicators whose activities are modulated by local transcription and chromatin architecture. Specific aims are: 1) to determine whether the human lamin B2 and [unreadable]-globin origins, which are thought to represent true replicators, are intrinsically more active than the most active sites in the DHFR initiation zone; ori-[unreadable] will be flanked with mismatched LoxP sites and exchanged with the two human origins, or with fragments that do not initiate efficiently in loco; the activities of each will then be quantified; lamin B2 and [unreadable]-globin also will be compared in human cells to fragments residing in initiation zones; 2) to determine whether ori-[unreadable] and ori-? are more active than other sites in the spacer because of their locations vis-[unreadable]-vis the local transcription units; ori-[unreadable] will be positioned differently in the spacer by the ROKO approach, and effects on origin efficiency assessed; 3) to determine whether DHFR and the weaker 2BE2121 promoters play essential but redundant roles in origin activation; both promoters will be deleted, and effects on origin activity will be assessed; in addition, a strong transcription stop signal will be placed just downstream from the DHFR promoter to determine whether transcription fork movement (as opposed to the promoter per se) is required for origin stimulation; 4) to determine why the efficiency of initiation increases when the intergenic spacer is truncated, and to prepare a "super" origin; the length of the spacer will be reduced to determine optimal size and a strong transcription terminator will be placed downstream from the DHFR and 2BE2121 genes; 5) to define the epigenetic changes that occur when ori-[unreadable] is inactivated by deletion of 5' and 3' regulatory elements or activated by one of the above strategies; ChIP technology will be utilized to determine the distributions of initiation proteins and modified histones as a function of the cell cycle in variants elaborated in Aims 1-4; these studies will indicate factors that limit origin efficiency, and will test the predictions that transcription through the spacer removes pre-RCs, while promoter deletions result in a failure to load them. [unreadable] [unreadable] [unreadable]